论文信息 - Minimal Experimental Size in the Three Way ANOVA Cross Classification Model with Approximate F-Tests

Minimal Experimental Size in the Three Way ANOVA Cross Classification Model with Approximate F-Tests

We consider a case of the three-way ANOVA where for testing the equality of the main effects of some fixed factor no exact F-test exists. For an approximate F-test the problem of finding the minimal size of the experiment so that given precision requirements are fulfilled is approximately solved. Our solutions have been found and tested by simulation experiments.

[1] R Core Team,et al. R: A language and environment for statistical computing. , 2014 .

[2] D. Rasch,et al. Determination of the size of a balanced experiment in mixed ANOVA models using the modified approximate F-test , 2005 .

[3] F. E. Satterthwaite. An approximate distribution of estimates of variance components. , 1946, Biometrics.

[4] Satterthwaite Fe. An approximate distribution of estimates of variance components. , 1946 .

[5] Guido Schwarzer,et al. Easier parallel computing in R with snowfall and sfCluster , 2009, R J..